Skip to main content
Version: 19.x (unreleased)

OAuth2 and OIDC Authentication

Report an issue with this page

This guide will explain how to configure an SSO provider using OpenID Connect (also known as OIDC) to issue Teleport credentials to specific groups of users. When used in combination with role-based access control (RBAC), OIDC allows Teleport administrators to define policies like:

  • Only members of the "DBA" group can connect to PostgreSQL databases.
  • Developers must never SSH into production servers.

How it works​

You can register your Teleport cluster as an application with your SSO provider, then create an authentication connector resource that provides Teleport with information about your application. When a user signs in to Teleport, your SSO provider executes its own authentication flow, then sends an HTTP request to your Teleport cluster to indicate that authentication has completed.

Teleport authenticates users to your infrastructure by issuing short-lived certificates. After a user completes an SSO authentication flow, Teleport issues short-lived TLS and SSH certificates to the user. Teleport also creates a temporary user on the Auth Service backend.

Teleport roles are encoded in the user's certificates. To assign Teleport roles to the user, the Auth Service inspects the role mapping within the authentication connector, which associates user data on your SSO provider with the names of one or more Teleport roles.

Prerequisites​

  • Admin access to the SSO/IdP being integrated with users assigned to groups/roles.
  • Teleport role with permission to maintain oidc resources. This permission is available in the default editor role.

  • A running Teleport Enterprise cluster. If you want to get started with Teleport, sign up for a free trial or set up a demo environment.

  • The tctl and tsh clients.

    Installing tctl and tsh clients

    1. Determine the version of your Teleport cluster. The tctl and tsh clients must be at most one major version behind your Teleport cluster version. Send a GET request to the Proxy Service at /v1/webapi/find and use a JSON query tool to obtain your cluster version. Replace teleport.example.com:443 with the web address of your Teleport Proxy Service:

      TELEPORT_DOMAIN=teleport.example.com:443
      TELEPORT_VERSION="$(curl -s https://$TELEPORT_DOMAIN/v1/webapi/find | jq -r '.server_version')"

    2. Follow the instructions for your platform to install tctl and tsh clients:

      Download the signed macOS .pkg installer for Teleport, which includes the tctl and tsh clients:

      curl -O https://cdn.teleport.dev/teleport-${TELEPORT_VERSION?}.pkg

      In Finder double-click the pkg file to begin installation.

      danger

      Using Homebrew to install Teleport is not supported. The Teleport package in Homebrew is not maintained by Teleport and we can't guarantee its reliability or security.

  • To check that you can connect to your Teleport cluster, sign in with tsh login, then verify that you can run tctl commands using your current credentials. For example, run the following command, assigning teleport.example.com to the domain name of the Teleport Proxy Service in your cluster and [email protected] to your Teleport username: 
    tsh login --proxy=teleport.example.com --user=[email protected]
    tctl status
    Cluster  teleport.example.com
    Version  19.0.0-dev
    CA pin   sha256:abdc1245efgh5678abdc1245efgh5678abdc1245efgh5678abdc1245efgh5678
    If you can connect to the cluster and run the tctl status command, you can use your current credentials to run subsequent tctl commands from your workstation. If you host your own Teleport cluster, you can also run tctl commands on the computer that hosts the Teleport Auth Service for full permissions.

Step 1/3. Configure your OIDC provider​

Register Teleport with the external identity provider you will be using and obtain your client_id and client_secret. This information should be documented on the identity providers website. Here are a few links:

Google Workspace

For Google Workspace, see Teleport Authentication with Google Workspace

Save the relevant information from your identity provider. To make following this guide easier, you can add the Client ID here and it will be included in the example commands below:

Client ID: CLIENT-ID

Select an OIDC redirect URL​

OIDC relies on HTTP redirects to return control back to Teleport after authentication is complete. The redirect URL must be selected by a Teleport administrator in advance.

The redirect URL for OIDC authentication in Teleport is mytenant.teleport.sh:443/v1/webapi/oidc/callback. Replace mytenant.teleport.sh:443 with your Teleport Cloud tenant or Proxy Service address. If you have a self-hosted cluster with multiple public addresses for the Teleport Proxy Service (the value of proxy_service.public_addr in the Teleport configuration file), ensure that this address points to the first one listed.

Step 2/3. Connect your OIDC provider to Teleport​

In this section, you will create an authentication connector that provides Teleport with the information it needs to exchange OIDC messages with your IdP and issue certificates to users.

Assign a role mapping​

When a user authenticates to Teleport, the Teleport Auth Service issues SSH and TLS certificates to the user that contain the user's Teleport roles.

For SSO authentication connectors, the Auth Service determines which roles to encode in the certificate by reading the authentication connector's role mapping. The role mapping indicates which Teleport roles to assign based on data that your identity provider stores about the user.

When you configure an authentication connector using the tctl CLI, a role mapping takes the following format:

<claim_name>,<claim_value>,<teleport_role_1>,<teleport_role_2>,...,<teleport_role_n>

For example, the following role mapping means that any user with the claim group with the value admin receives the Teleport roles editor and auditor:

group,admin,editor,auditor

For the purpose of this guide, assign two separate role mappings:

  • A more permissive role mapping: group,admins,auditor,editor
  • A more restrictive role mapping: group,users,access

Configure an OIDC connector​

The next step is to add an OIDC connector to Teleport. The connectors are created, tested, and added or removed using tctl resource commands or the Teleport Web UI.

On your workstation, create a file called client-secret.txt consisting only of your client secret.

To create a new connector, use tctl sso configure. The following example creates a connector resource file in YAML format named oidc-connector.yaml:

tctl sso configure oidc --name <CONNECTOR-NAME> \  --issuer-url <PATH-TO-PROVIDER> \  --id CLIENT-ID \  --secret $(cat client-secret.txt) \  --claims-to-roles mapping_1 \  --claims-to-roles mapping_2 > oidc-connector.yaml
  • --name: Usually the name of the IdP, this is how the connector will be identified in Teleport.
  • --issuer-url: This is the base path to the IdP's OIDC configuration endpoint, excluding .well-known/openid-configuration. If, for example, the endpoint is https://example.com/.well-known/openid-configuration, you would use https://example.com.
  • --id: The client ID as defined in the IdP. Depending on your identity provider this may be something you can define (for example, teleport), or may be an assigned string.
  • --secret: The client token/secret provided by the IdP to authorize this client.

For more information on these and all available flags, see the tctl sso configure oidc section of the Teleport CLI Reference page.

The file created should look like the example below:

kind: oidc
metadata:
  name: oidc_connector
spec:
  claims_to_roles:
  - claim: groups
    roles:
    - access
    value: users
  - claim: groups
    roles:
    - editor
    value: admins
  client_id: <CLIENT-NAME>
  client_secret: <CLIENT-SECRET>
  issuer_url: https://idp.example.com/
  redirect_url: https://mytenant.teleport.sh:443/v1/webapi/oidc/callback
  max_age: 24h
  # pkce_mode determines if the OIDC authentication flow should include PKCE code verifiers. Options or `enabled`, `disabled`, and `` (defaults to disabled)
  pkce_mode: "enabled"
  client_redirect_settings:
    # a list of hostnames allowed for HTTPS client redirect URLs
    # can be a regex pattern
    allowed_https_hostnames:
      - remote.machine
      - '*.app.github.dev'
      - '^\d+-[a-zA-Z0-9]+\.foo.internal$'
    # a list of CIDRs allowed for HTTP or HTTPS client redirect URLs
    insecure_allowed_cidr_ranges:
      - '192.168.1.0/24'
      - '2001:db8::/96'
version: v3
Details

Practical Example: Keycloak The following example was generated using Keycloak as the identity provider. Keycloak is being served at keycloak.example.com, and the Teleport Proxy Service is listening at teleport.example.com. In Keycloak, the client is named teleport. Under the teleport-dedicated client scope, we've added the "Group Membership" mapper:

kind: oidc
metadata:
  name: keycloak
spec:
  claims_to_roles:
  - claim: groups
    roles:
    - access
    value: /users
  - claim: groups
    roles:
    - editor
    value: /admins
  client_id: teleport
  client_secret: abc123...
  issuer_url: https://keycloak.example.com/realms/master
  redirect_url: https://teleport.example.com:443/v1/webapi/oidc/callback
version: v3

Optional: ACR Values​

Teleport supports sending Authentication Context Class Reference (ACR) values when obtaining an authorization code from an OIDC provider. By default ACR values are not set. However, if the acr_values field is set, Teleport expects to receive the same value in the acr claim, otherwise it will consider the callback invalid.

In addition, Teleport supports OIDC provider specific ACR value processing which can be enabled by setting the provider field in OIDC configuration. At the moment, the only build-in support is for NetIQ.

A example of using ACR values and provider specific processing is below:

# example connector which uses ACR values
kind: oidc
version: v2
metadata:
  name: "oidc-connector"
spec:
  issuer_url: "https://oidc.example.com"
  client_id: "xxxxxxxxxxxxxxxxxxxxxxx.example.com"
  client_secret: "zzzzzzzzzzzzzzzzzzzzzzzz"
  redirect_url: "https://mytenant.teleport.sh/v1/webapi/oidc/callback"
  display: "Login with Example"
  acr_values: "foo/bar"
  provider: netiq
  scope: [ "group" ]
  claims_to_roles:
     - claim: "group"
       value: "editor"
       roles: [ "editor" ]
     - claim: "group"
       value: "user"
       roles: [ "access" ]

Optional: Max age​

The max_age field controls the maximum age of users' sessions before they will be forced to reauthenticate. By default max_age is unset, meaning once a user authenticates using OIDC they will not have to reauthenticate unless the configured OIDC provider forces them to. This can be set to a duration of time to force users to reauthenticate more often. If max_age is set to zero seconds, users will be forced to reauthenticate with their OIDC provider every time they authenticate with Teleport.

Note that the specified duration must be in whole seconds. 24h works because that's the same as 1440s, but 60s500ms would not be allowed as that is 60.5 seconds.

# Extra parts of OIDC yaml have been removed.
spec:
  max_age: 24h

Note that not all OIDC providers support setting max_age. Google and GitLab are both known not to support it and authentication with those providers will not work when the max_age field is set.

Optional: Prompt​

Set the Authorization Server prompt for the End-User for reauthentication and consent per the OIDC protocol. If no prompt value is set, Teleport uses select_account as default.

# Extra parts of OIDC yaml have been removed.
spec:
  # Valid values as defined from https://openid.net/specs/openid-connect-core-1_0.html#AuthRequest
  # none: The Authorization Server must not display any authentication or consent user interface pages.
  # select_account: The Authorization Server should prompt the End-User to select a user account.
  # login: The Authorization Server should prompt the End-User for reauthentication.
  # consent: The Authorization Server should prompt the End-User for consent before returning information to the Client.
  prompt: 'login'

Optional: Redirect URL and Timeout​

The redirect URL must be accessible by all user, optional redirect timeout.

# Extra parts of OIDC yaml have been removed.
spec:
  redirect_url: https://<cluster-url>.example.com:3080/v1/webapi/oidc/callback
  # Optional Redirect Timeout.
  # redirect_timeout: 90s

Optional: Disable email verification​

By default, Teleport validates the email_verified claim, and users who attempt to sign in without a verified email address are prevented from doing so:

ERROR: SSO flow failed.
identity provider callback failed with error: OIDC provider did not verify email.
        email not verified by OIDC provider

For testing and other purposes, you can opt out of this behavior by enabling allow_unverified_email in your OIDC connector. This option weakens the overall security of the system, so we do not recommend enabling it.

kind: oidc
version: v2
metadata:
  name: connector
spec:
  allow_unverified_email: true

Optional: Specify a claim to use as the username​

By default, Teleport will use the user's email as their Teleport username.

You can define a username_claim to specify the claim that should be used as the username instead:

kind: oidc
version: v2
metadata:
  name: connector
spec:
  # Use the `preferred_username` claim as the user's Teleport username.
  username_claim: preferred_username

Optional: Request object mode​

Teleport supports the ability to send authorization request parameters in a JSON Web Token (JWT), referred to as "request objects", following RFC 9101. Presently, Teleport only supports signed request objects sent by value. This option necessitates that the IdP be provided with, or configured to discover, Teleport's JSON Web Key Sets (JWKS) used for OIDC IdP integration in order to verify request object signatures. This JWKS can be found at the /.well-known/jwks-oidc path of the Web API.

Additionally, the public key to be used for verifying request object signatures can be found via tctl.

tctl get cert_authority/oidc_idp/$CLUSTER_NAME --format=json | jq -r .[].spec.active_keys.jwt[].public_key

By default, request_object_mode is unset, meaning authorization request parameters will be sent in the query string of the authorization endpoint.

kind: oidc
version: v2
metadata:
  name: connector
spec:
  # Use signed request objects when making authorization requests to the IdP.
  request_object_mode: signed

Request objects are also supported for MFA checks. If request_object_mode is not explicitly set on the MFA client, the login client's request object mode will be used by default. If desired, you can explicitly configure the MFA client to use a separate request object mode.

kind: oidc
version: v2
metadata:
  name: connector
spec:
  client_id: teleport_login
  client_secret: abc123...
  request_object_mode: none
  mfa:
    client_id: teleport_mfa
    client_secret: mfa123...
    request_object_mode: signed

Note: Support for request objects is available in Teleport Enterprise versions 17.7.2 or later and 18.1.6 or later.

Test the connector​

Before applying the connector to your cluster, you can test that it's configured correctly:

cat oidc-connector.yaml | tctl sso test

This should open up your web browser and attempt to log you in to the Teleport cluster through your IdP. If it fails, review the output of this command for troubleshooting details.

tip

The "[OIDC] Claims" section of the CLI output provides all the details of your user provided by the IdP. This is a good starting point while troubleshooting errors like Failed to calculate user attributes.

Create the connector​

After your tests are successful, create the connector:

tctl create -f oidc-connector.yaml

Step 3/3. Enable default OIDC authentication​

Edit your cluster authentication preferences so you can make the authentication connector you configured in this guide the default authentication method for your Teleport cluster.

Open the Teleport Web UI. From the left sidebar, navigate to Zero Trust Access > Auth Connectors. Find the connector you would like to make the default and, from its three-dot menu, select Set as default.

If you are managing your Teleport resources as configuration files, you can choose a default authentication connector using a dynamic resource. In this case, use tctl to edit the cluster_auth_preference value:

tctl edit cluster_auth_preference

Set the value of spec.type to oidc:

kind: cluster_auth_preference
metadata:
  ...
  name: cluster-auth-preference
spec:
  ...
  type: oidc
  ...
version: v2

After you save and exit the editor, tctl will update the resource:

cluster auth preference has been updated
Additional ways to edit cluster auth preferences

The cluster auth preference is available as a Teleport Terraform provider resource. Find a list of configuration options in the Cluster Auth Preferences Resource Reference.

If you self-host Teleport, you can edit your Teleport Auth Service configuration file to include the following:

# Snippet from /etc/teleport.yaml
auth_service:
  authentication:
    type: oidc

If you need to log in again before configuring your identity provider, use the flag --auth=local.

Next steps​

Now that you have connected Teleport to your identity provider, you can customize the way Teleport includes IdP data in Teleport roles.

With role templates, you can include user data from your IdP directly in Teleport roles. If you use the external template variable in the value of a role field, Teleport passes that value from your IdP. In this example, all of the role options you can use for allowing users to assume certain principals on remote systems come from your IdP:

kind: role
version: v7
metadata:
  name: sso-users
spec:
  allow:
    logins: ['{{external.logins}}']
    aws_role_arns: ['{{external.aws_role_arns}}']
    azure_identities: ['{{external.azure_identities}}']
    db_names: ['{{external.db_names}}']
    db_roles: ['{{external.db_roles}}']
    db_users: ['{{external.db_users}}']
    desktop_groups: ['{{external.desktop_groups}}']
    gcp_service_accounts: ['{{external.gcp_service_accounts}}']
    host_groups: ['{{external.host_groups}}']
    host_sudoers: ['{{external.host_sudoers}}']
    kubernetes_groups: ['{{external.kubernetes_groups}}']
    kubernetes_users: ['{{external.kubernetes_users}}']
    windows_desktop_logins: ['{{external.windows_desktop_logins}}']

For more information on using the external template variable, see Role Templates.

For an explanation of the fields listed above, see the Role Reference.

If you need to transform your IdP user data before you include it in Teleport roles, you can do so using Login Rules. Login Rules allow you to include external traits within Teleport roles even if your IdP provides user data in a different format than the one expected by Teleport. Read more about Login Rules.

Troubleshooting​

Troubleshooting SSO configuration can be challenging. Usually a Teleport administrator must be able to:

  • Be able to see what SAML/OIDC claims and values are getting exported and passed by the SSO provider to Teleport.
  • Be able to see how Teleport maps the received claims to role mappings as defined in the connector.
  • For self-hosted Teleport Enterprise clusters, ensure that HTTP/TLS certificates are configured properly for both the Teleport Proxy Service and the SSO provider.

If something is not working, we recommend to:

  • Double-check the host names, tokens and TCP ports in a connector definition.

Using the Web UI​

If you get "access denied" or other login errors, the number one place to check is the Audit Log. To view the recording, select Audit in the Teleport Web UI, then click Session Recordings in the menu.

Example of a user being denied because the role clusteradmin wasn't set up:

{
  "code": "T1001W",
  "error": "role clusteradmin is not found",
  "event": "user.login",
  "message": "Failed to calculate user attributes.\n\trole clusteradmin is not found",
  "method": "oidc",
  "success": false,
  "time": "2024-11-07T15:41:25.584Z",
  "uid": "71e46f17-d611-48bb-bf5e-effd90016c13"
}

Teleport does not show the expected Nodes​

When the Teleport Auth Service receives a request to list Teleport-connected resources (e.g., to display resources in the Web UI or via tsh ls), it only returns the resources that the current user is authorized to view.

For each resource in the user's Teleport cluster, the Auth Service applies the following checks in order and, if one check fails, hides the resource from the user:

  • None of the user's roles contain a deny rule that matches the resource's labels.
  • At least one of the user's roles contains an allow rule that matches the resource's labels.

If you are not seeing resources when expected, make sure that your user's roles include the appropriate allow and deny rules as documented in the Access Controls Reference.

When configuring SSO, ensure that the identity provider is populating each user's traits correctly. For a user to see a Node in Teleport, the result of populating a template variable in a role's allow.logins must match at least one of a user's traits.logins.

In this example a user will have usernames ubuntu, debian and usernames from the SSO trait logins for Nodes that have a env: dev label. If the SSO trait username is bob then the usernames would include ubuntu, debian, and bob.

kind: role
metadata:
  name: example-role
spec:
  allow:
    logins: ['{{external.logins}}', ubuntu, debian]
    node_labels:
      'env': 'dev'
version: v5

Single sign-on fails with OIDC​

When encountering the error message "Failed to verify JWT: oidc: unable to verify JWT signature: no matching keys", it typically indicates a discrepancy between the algorithm used to sign the JWT token and the algorithm(s) supported by the JSON Web Key Set (JWKS). Specifically, the token might be signed with one algorithm, e.g., HS256, while the JWKS only lists keys for a different algorithm. e.g., RS256. This issue predominantly arises when using identity providers that offer extremely low-level functionality.

Here are some things to check:

  • Verify the JWT header specifies the correct signing algorithm. This should match one of the algorithms listed in the keys section of the JWKS endpoint response.
  • Ensure the JWKS endpoint is returning all relevant public keys. Sometimes key rotation can cause valid keys to be omitted.

To resolve the issue, align the JWT algorithm header with a supported algorithm in the JWKS. Rotate keys if necessary. Verify the JWKS only publishes the active public keys. With proper configuration, the signature should validate successfully.